Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

integrate image processing algorithms to detect foliage area #169

Open
jakerye opened this issue Oct 31, 2018 · 2 comments
Open

integrate image processing algorithms to detect foliage area #169

jakerye opened this issue Oct 31, 2018 · 2 comments
Labels
ARCHIVE These are out of date. enhancement New feature or request

Comments

@jakerye
Copy link
Contributor

jakerye commented Oct 31, 2018

This is challenging to get reliable but it would be great to implement some of the "demo" algorithms into the code base.

There has already been work that detects number of leaves, plant width, etc.

There needs to be some pre-processing filters to determine if an image is valid to be processed. The main concern is to detect occlusion (what happens when the plants get big and they overlap?) Can we put the plant site locations in the device config file? Then make sure the image has isolated clusters around each plant site. Then if there is not occlusion, run through the leaf count processor. The leaf count processor will likely need to know the cultivar classification (Basil leaves vs. Arugula leaves).

@jakerye jakerye added the enhancement New feature or request label Oct 31, 2018
@jakerye jakerye changed the title integrate image processing algorithms to detect low hanging fruit integrate image processing algorithms to detect foliage area Oct 31, 2018
@jakerye
Copy link
Contributor Author

jakerye commented Oct 31, 2018

See work done previously by Eddie

@ghost
Copy link

ghost commented Oct 31, 2018

Maybe you could tag the plant sites with some kind of unique fiducial marker like a 2D barcode. That might also help with identifying cameras for issue #168 (multiple cameras). Eventually the markers would get occluded, but you could identify the sites at transplant time. You could even take it so far as to have stickers or snap in labels to uniquely identify each seedling (traceable back to cultivar classification, seed provenance, etc.)

Here's a tutorial on using OpenCV to detect aruco markers: https://docs.opencv.org/3.1.0/d5/dae/tutorial_aruco_detection.html

Here's one for generating aruco markers in python: https://mecaruco2.readthedocs.io/en/latest/notebooks_rst/Aruco/aruco_basics.html

@jakerye jakerye added this to the Edge Computing milestone Jan 30, 2019
@srmoore srmoore added the ARCHIVE These are out of date. label Jan 14, 2020
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
ARCHIVE These are out of date. enhancement New feature or request
Projects
None yet
Development

No branches or pull requests

2 participants